Free Radical Biology & Medicine 43 (2007) 695–701www.elsevier.com/locate/freeradbiomed

Original Contribution

In vivo interconversion of 7β-hydroxycholesterol and 7-ketocholesterol,potential surrogate markers for oxidative stress

Hanna Larsson a, Ylva Böttiger b, Luigi Iuliano c, Ulf Diczfalusy a,⁎

a Division of Clinical Chemistry, Department of Laboratory Medicine, Karolinska Institutet, Huddinge, SE-141 86 Stockholm, Swedenb Clinical Pharmacology, Department of Laboratory Medicine, Karolinska Institutet, Huddinge, SE-141 86 Stockholm, Sweden

c Department of Internal Medicine, University La Sapienza, Rome, Italy

Received 2 March 2007; revised 16 April 2007; accepted 28 April 2007Available online 10 May 2007

Abstract

The oxysterols 7β-hydroxycholesterol and 7-ketocholesterol are cholesterol autoxidation products. These two oxysterols are formed as a resultof low density lipoprotein oxidation and in a study on biomarkers for oxidative stress in patients with atherosclerosis, 7β-hydroxycholesterol wasfound to be the strongest predictor of progression of carotid atherosclerosis. Interconversion of 7β-hydroxycholesterol and 7-ketocholesterol invitro has been reported recently, using recombinant 11β-hydroxysteroid dehydrogenase or rodent liver microsomes. In this study deuterium-labeled 7β-hydroxycholesterol or 7-ketocholesterol was administered intravenously to two healthy volunteers and blood samples were collected atdifferent time points. The mean half-life for elimination of 7β-hydroxycholesterol from the circulation was estimated to be 1.9 h. Thecorresponding half-life for 7-ketocholesterol was estimated to be 1.5 h. Infusion of deuterium-labeled 7-ketocholesterol resulted in labeling of 7β-hydroxycholesterol and vice versa. In addition, the biological within-day and between-day variations of the two oxysterols were determined. Insummary, the present investigation clearly shows an interconversion of 7β-hydroxycholesterol and 7-ketocholesterol in humans.© 2007 Elsevier Inc. All rights reserved.

Keywords: 11β-Hydroxysteroid dehydrogenase; Oxysterols; Cholesterol oxidation products; Half-life

Introduction

Cholesterol is sensitive to oxygen and easily oxidizes to theoxysterols 7β-hydroxycholesterol and 7-ketocholesterol [1].These autoxidation products are also formed during in vitrooxidation of low density lipoprotein [2,3] and are majoroxysterols in atherosclerotic plaques [4]. The two oxysterolshave therefore been suggested as surrogate markers for in vivolipoprotein oxidation [5,6] and oxidative stress in general [7,8].

However, there are no data on the normal intraindividualvariation in plasma concentration of these oxysterols. Further-more, different authors have used either 7β-hydroxycholesterolor 7-ketocholesterol as an oxidative stress marker assuming thatthe two markers are formed independently. Interconversion of7β-hydroxycholesterol and 7-ketocholesterol in rat liver was

Abbreviations: BHT, butylated hydroxytoluene; CYP, cytochrome P450.⁎ Corresponding author. Fax: +46 8 585 812 60.E-mail address: ulf.diczfalusy@karolinska.se (U. Diczfalusy).

0891-5849/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.freeradbiomed.2007.04.033

reported by Björkhem et al. already in 1968 [9], although theenzyme(s) responsible for the conversions were not identified.Furthermore, high concentrations of 7β-hydroxycholesterolwere found in the liver of rats after infusion of a fat emulsioncontaining 7-ketocholesterol [10]. A rapid hepatic conversion of7-ketocholesterol to a polar metabolite, tentatively identified as7β-hydroxycholesterol, was observed following injection ofradiolabeled 7-ketocholesterol in rats [11,12]. In addition,apparent conversion of radiolabeled 7-ketocholesterol into 7β-hydroxycholesterol has also been reported in cultured humanmacrophages [13] and the human hepatoma cell line HepG2[14]. In 1994 Song et al. reported on a 7α-hydroxycholesteroloxidoreductase activity found in livers from humans andhamsters [15]. The enzyme from hamster liver was furthercharacterized [16,17] and shown to be similar to type 1 11β-hydroxysteroid dehydrogenase. Subsequent molecular cloningof the 11β-hydroxysteroid dehydrogenase type 1 enzyme fromrat, mouse, hamster, and human showed that this enzymemediates the interconversion of 7-ketocholesterol to 7-hydro-

696 H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

xycholesterol [18,19]. The enzyme from hamster, however, actsas both a 7α- and a 7β-hydroxycholesterol dehydrogenasewhile the human and rodent enzymes stereospecifically catalyzethe conversion of 7β-hydroxycholesterol to 7-ketocholesterol.

The present investigation was undertaken to determine thewithin-day and between-day variation in plasma concentrationof 7β-hydroxycholesterol and 7-ketocholesterol in 12 healthysubjects and to find out whether interconversion of 7-ketocholesterol and 7β-hydroxycholesterol, previously demon-strated in vitro and in experimental animals, also takes place invivo in humans.

Materials and methods

Materials

[25, 26, 26, 26, 27, 27, 27- 2H7]7-Ketocholesterol waspurchased from CDN Isotopes (Pointe-Claire, Quebec, Canada).[26,26,26,27,27,27-2H6]7β-Hydroxycholesterol was synthesizedas described in [20]. 7-Ketocholesterol, 7β-hydroxycholesterol,25-hydroxycholesterol, and butylated hydroxytoluene (BHT)were from Sigma (St. Louis, MO). 7α-Hydroxycholesterol, 27-hydroxycholesterol, and cholestan-3β,5α,6β-triol were fromResearch Plus Inc. (Manasquan, NJ). 24(S)-Hydroxycholesterolwas purchased from Steraloids Inc. (Newport, RI). Humanalbumin 200 g/L was purchased from Octapharma (Stockholm,Sweden). Potassium hydroxide, phosphoric acid, EDTA, andtrimethylsilylchlorosilane were from VWR (Stockholm, Swe-den). Hexamethyldisilazane and pyridine were purchased fromPierce (Rockford, IL). Isolute SI solid-phase extraction columns100 mg/1 ml were obtained from Sorbent AB (Västra Frölunda,Sweden).

Determination of intraindividual variations in plasmaoxysterol concentrations

Study subjectsThe intraindividual variation in plasma concentration of 7β-

hydroxycholesterol and 7-ketocholesterol was determined inhealthy volunteers. For comparison, concentrations of 7α-hydroxycholesterol, cholestanetriol (5α-cholestane-3β,5,6β-triol), 24-hydroxycholesterol, 25-hydroxycholesterol, and 27-hydroxycholesterol were also determined.

Twelve subjects were included, 7 women and 5 men, andthey were all healthy according to a health declaration.Demographics of the subjects are specified in Table 1. Noneof the subjects abused alcohol, smoked, or took illicit drugs. Thesubjects were not on any regular drug treatment, except subject1, who took levaxin 0.1 mg daily, and subject 11, who took

Table 1Demographic data on subjects participating in the study on intraindividual variation

Number of subjects Age (years)

Women 7 39 (23–63)Men 5 39 (28–57)

Values expressed as mean (range).

combined oral contraceptives. None of the women werepregnant or lactating during the study period.

Sampling proceduresSampling (10 ml of whole blood) on Day 1 was performed

from an indwelling catheter in the cubital vein at 08.00, 12.00,16.00, and 20.00 o'clock, and by venipuncture at approximately08.00 Days 2–5, Day 8, Day 15, Day 22, and after 3 months.Blood samples were collected in tubes containing K2EDTA.Samples were immediately centrifuged at 1300g for 12 min, andplasma was separated and stored at −70°C until analysis.

Analysis of oxysteroidsOxysterols were determined by combined gas chromatogra-

phy–mass spectrometry using deuterium-labeled internal stan-dards, as described by Dzeletovic et al. [20]. All samples fromeach subject were analyzed at one occasion.

Infusion of deuterium-labeled 7-ketocholesterol and7β-hydroxycholesterol

Infusion and sampling proceduresDeuterium-labeled 7-ketocholesterol (240 μg) in human

serum albumin containing 10% ethanol (0.5 ml) was admini-strated intravenously to a healthy male volunteer 65 years of age(Volunteer A, body mass index=26.9, weight 88 kg) and 300 μgdeuterium-labeled 7-ketocholesterol was given to a healthyfemale volunteer 69 years of age (Volunteer B, body massindex=25.2, weight 67 kg). Blood samples were collectedbefore and at specific time points after administration in tubescontaining K2EDTA. Samples were immediately centrifuged at1300g for 12 min, and plasma was separated and stored at−70°C until analysis. Deuterium-labeled 7β-hydroxycholes-terol (300 μg) was administrated under the same conditions asdescribed above to the same two healthy volunteers. Bloodsamples were collected and stored as described above.

AnalysisPlasma, containing 200 μg EDTA and 50 μg BHT per

milliliter, was transferred to a screw-capped glass reaction vialwith a Teflon-lined septum. Argon was flushed through the vialfor 20 min before 0.35 M ethanolic potassium hydroxide wasadded. The sample was kept at room temperature with magneticstirring for 2 h. After that the sample was transferred withchloroform to a separatory funnel containing 0.15 M sodiumchloride, and pH was adjusted to 7 with phosphoric acid. Aftervigorous shaking the organic phase was transferred to a conicalbottom flask and evaporated using a rotary evaporator. Theresidue was dissolved in toluene and put on an Isolute SI solid-

in plasma oxysterol concentrations

Weight (kg) Height (cm) BMI (kg/m2)

60 (55–66) 167 (160–173) 21.5 (19.0–24.2)71 (62–85) 180 (168–189) 21.9 (19.1–25.0)

Fig. 1. Variations in plasma oxysterol concentrations (ng/ml) with time in onehealthy volunteer. 7a-OH, 7α-hydroxycholesterol; 7b-OH, 7β-hydroxycholesterol;7-keto, 7-ketocholesterol; 3,5,6-triol, cholestane-3β,5α,6β-triol; 24-OH, 24-hydro-xycholesterol; 25-OH, 25-hydroxycholesterol; 27-OH, 27-hydroxycholesterol.

697H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

phase extraction column preconditioned with hexane. Thecolumn was washed with hexane and 0.5% 2-propanol inhexane before the sample was eluted with 30% 2-propanol inhexane. The solvent was gently evaporated using nitrogenbefore a derivatization mixture containing pyridine/hexam-ethyldisilazane/trimethylchlorosilane 3/2/1 (v/v/v) was added tothe sample. The sample was incubated at 60°C for 30 minbefore the solvent was evaporated using nitrogen. The residuewas dissolved in hexane and analyzed by gas chromatography–mass spectrometry.

Gas chromatography–mass spectrometry was performed ona Hewlett Packard 5890 gas chromatograph equipped with aHP-5MS capillary column (30 m×0.25 mm, 0.25 μm phasethickness) connected to a Hewlett Packard 5970 massspectrometer. Helium was used as carrier gas at a flow rateof 0.8 ml/min. The injector temperature was 270°C and thedetector temperature 260°C. The oven temperature programwas as follows: An initial temperature of 180°C was held for1 min before the temperature was increased by 20°C perminute to 250°C. From there the temperature was increased by4°C per minute to 300°C where it was kept for 8 min. Themass spectrometer was used in selected ion monitoring mode.After infusion of deuterium-labeled 7-ketocholesterol thefollowing ions (m/z) were monitored: 479 ([2H7]7-ketocho-lesterol), 463 ([2H7]7β-hydroxycholesterol), 463 ([2H7]7α-hydroxycholesterol), 456 (7β-hydroxycholesterol), 456 (7α-hydroxycholesterol), and 472 (7-ketocholesterol). After infu-sion of [2H6]7β-hydroxycholesterol the following ions (m/z)were monitored: 478 ([2H6]7-ketocholesterol), 462 ([2H6]7β-hydroxycholesterol), 462 ([2H6]7α–hydroxycholesterol), 456(7β-hydroxycholesterol), 456 (7α-hydroxycholesterol), and472 (7-ketocholesterol). The electron ionization energy was70 eV.

Ethical aspects

All subjects gave a written, informed consent to participate inthis study, which was approved by the Ethics committee ofKarolinska Institutet at Huddinge University Hospital, Hud-dinge, Sweden.

Results

Determination of the intraindividual variation in plasmaconcentration of 7β-hydroxycholesterol and 7-ketocholesterol

The plasma concentrationsof several oxysterols were deter-mined at 12 different time points within a 3-month period. Themean value for the 12 measurements in each subject wascalculated, as was the standard deviation and the relativestandard variation (CV%). The results from one of the volunteersare shown as an example in Fig. 1. In Table 2 the mean plasmaconcentrations of the oxysterols as well as the relative standarddeviations from the 12 subjects are given. The oxysterol 7β-hydroxycholesterol showed a larger intraindividual variationwith time than 7-ketocholesterol. For comparative purposes, thecorresponding variations were calculated also for 7α-hydro-

xycholesterol, cholestane-3β,5α,6β-triol, 24-hydroxycholes-terol, 25-hydroxycholesterol, and 27-hydroxycholesterol. Asshown in Table 2 the variation for 24-hydroxycholesterol and27-hydroxycholesterol was lower than for all other oxysterolsdetermined. These two oxysterols are formed enzymatically bythe cytochrome P450 (CYP) enzymes CYP46 and CYP27A1,respectively [21]. The biological variation in oxysterol con-centration in the volunteers should be compared to themethodological variation. Variations in oxysterol concentrationsin 12 identical plasma control samples prepared individuallyduring one day and analyzed in one batch are shown in Table 3. Itis evident that there is a large biological variation in plasmaoxysterol concentrations with time.

Infusion experiments with deuterium-labeled 7-ketocholesteroland 7β-hydroxycholesterol

Infusion of deuterium-labeled 7-ketocholesterolTo study the elimination kinetics of 7-ketocholesterol from

the human circulation deuterium-labeled 7-ketocholesterol was

Fig. 2. Elimination of injected [2H7]7-ketocholesterol from the circulation in twohealthy volunteers. The amount of 240 μg [2H7]7-ketocholesterol wasadministered to volunteer A and 300 μg to volunteer B. The elimination of[2H7]7-ketocholesterol followed first-order kinetics and the half-life wascalculated to be ∼1.5 h.

Table 2Seven different oxysterols were determined in plasma from 12 healthyvolunteers at 12 occasions during a 3-month period

Oxysterol 7a-OH 7b-OH 7-Keto triol 24-OH 25-OH 27-OHMean 45.7 3.3 12.5 4.8 60.2 3.7 130.2SD 11.4 1.0 2.3 1.4 4.9 0.6 12.5CVMin 12.5 9.4 11.3 11.5 4.4 4.4 6.3CVMax 66.4 98.7 41.4 51.8 13.3 45.3 14.3Intraindividual CV 29.1 32.3 19.3 25.5 7.9 18.6 9.8Interindividual CV 65.4 26.1 19.6 60.1 20.7 35.2 19.7

The mean values of the oxysterol concentrations in each volunteer werecalculated as well as the standard deviations and the relative standard deviations(intraindividual CV, CV=SD/mean∗100). The averaged values of the 12individual mean values and standard deviations are shown in the table. Thehighest (CVMax) and lowest (CVMin) individual CVs among the 12 healthyvolunteers are shown for each oxysterol. The interindividual CVs are alsoshown for comparison. 7a-OH, 7α-hydroxycholesterol; 7b-OH, 7β-hydro-xycholesterol; 7-keto, 7-ketocholesterol; 3,5,6-triol, cholestane-3β,5α,6β-triol;24-OH, 24-hydroxycholesterol; 25-OH, 25-hydroxycholesterol; 27-OH,27-hydroxycholesterol.

698 H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

administered intravenously to two healthy volunteers. Asshown in Fig. 2 the deuterium label in the circulating 7-ketocholesterol increased rapidly and the labeled fractionreached 33 and 53% within 6 min in the male (A) and female(B) volunteer, respectively. The elimination followed first-orderkinetics and the half-life was estimated to be 1.3 and 1.7 h in thetwo volunteers. The apparent pool size of unlabeled 7-ketocholesterol was estimated from the 2H7/(

2H0+2H7) ratios

in the two volunteers and was found to be 0.8 and 0.6 mg in themale and female volunteer, respectively.

The two oxysterols 7α- and 7β-hydroxycholesterol werealso monitored during the infusion experiments, and it wasfound that deuterium label appeared in circulating 7β-hydro-xycholesterol corresponding to 2–3% of the total 7β-hydro-xycholesterol in the circulation (Fig. 3). There was however noincorporation of deuterium label in 7α-hydroxycholesterol.

Infusion of deuterium-labeled 7β-hydroxycholesterolThe elimination of 7β-hydroxycholesterol from the human

circulation was studied by administrating deuterium-labeled7β-hydroxycholesterol intravenously to two healthy volunteers.The deuterium-labeled fraction of 7β-hydroxycholesterolreached 82 and 78% within 5 min in the two volunteers,respectively. The half-life of 7β-hydroxycholesterol wasestimated from the regression lines for the elimination (Fig. 4)to be 1.8 and 2.0 h for the female and male volunteer,

Table 3Twelve aliquots of a control sample were prepared and analyzed individuallyduring one day

Oxysterol 7a-OH 7b-OH 7-Keto triol 24-OH 25-OH 27-OHMean 81.47 5.42 14.67 4.08 51.56 5.45 129.77SD 3.05 0.74 1.27 0.76 2.51 0.42 11.48CV 3.74 13.73 8.64 18.66 4.86 7.71 8.85

The mean concentrations (ng/ml), standard deviation (SD), and relative standarddeviation (SD/mean∗100, CV) of seven different oxysterols are tabulated. 7a-OH, 7α-hydroxycholesterol; 7b-OH, 7β-hydroxycholesterol; 7-keto, 7-keto-cholesterol; 3,5,6-triol, cholestane-3β,5α,6β-triol; 24-OH, 24-hydroxycholes-terol; 25-OH, 25-hydroxycholesterol; 27-OH, 27-hydroxycholesterol.

respectively. The apparent pool size of unlabeled 7β-hydro-xycholesterol was estimated to be 0.07 and 0.06 mg in the twovolunteers, respectively.

Infusion of deuterium-labeled 7β-hydroxycholesterol resultedin a rapid incorporation of deuterium label in circulating7-ketocholesterol, as shown in Fig. 5.

Discussion

According to the oxidation hypothesis, oxidative modifica-tion of low density lipoprotein (LDL) increases its atherogeni-city [22–24]. In order to monitor treatment in atheroscleroticpatients or to detect patients at risk of developing athero-sclerosis a clinical marker of lipoprotein oxidation would bevaluable. During in vitro oxidation of LDL cholesterol isoxidized to a number of oxidation products known asoxysterols [2,3]. High concentrations of oxysterols, at least100-fold higher than in human plasma, are found in humanatherosclerotic plaques [4,25]. Oxysterols have therefore beenstudied as potential markers of in vivo lipoprotein oxidation. Ina study of early carotid atherosclerosis in hypercholesterolemicFinnish men, an association between lipid oxidation andatherogenesis was found. The oxysterol 7β-hydroxycholesterolwas shown to be the strongest single predictor of progression of

Fig. 3. Appearance of deuterium label in circulating 7β-hydroxycholesterol afterinjection of [2H7]7-ketocholesterol into two healthy volunteers. Within 2 h afterinfusion the corresponding deuterium label appeared in 2–3% of totalcirculating 7β-hydroxycholesterol.

Table 4Half-lives of some oxysterols in the human circulation

Oxysterol t1/2 (h) Reference

7α-Hydroxycholesterol b0.5 [40]27-Hydroxycholesterol b0.8 [40]7-Ketocholesterol 1.5 This study3β-Hydroxy-5-cholestenoic acid 1.8 [41]7β-Hydroxycholesterol 1.9 This study24-Hydroxycholesterol 12 [42]4β-Hydroxycholesterol 62 [34]

Fig. 4. Elimination of injected [2H6]7β-hydroxycholesterol from the circulationin two healthy volunteers. The amount of 300 μg [2H6]7β-hydroxycholesterolwas administered to volunteers A and B. The elimination of [2H6]7β-hydroxycholesterol followed first-order kinetics and the half-life was calculatedto be ∼1.9 h.

699H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

carotid atherosclerosis [26]. Increased plasma levels of 7β-hydroxycholesterol were reported in a population with anincreased risk for cardiovascular disease [27]. In another studywhere long-term supplementation of vitamin E, vitamin C, or acombination of vitamin E+C was given to 48 male participantsit was shown that a daily supplement of 182 mg of vitamin E ora daily supplement of 182 mg of vitamin E+500 mg vitamin Cfor 1 year reduced serum 7β-hydroxycholesterol by 50%compared to placebo [28]. Oxysterols have also been suggestedas markers for oxidative stress in a wider perspective. Smokinghas been suggested to lead to increased oxidative stress, forexample, as determined by plasma malondialdehyde [29]. Theplasma level of 7-ketocholesterol was reported to be signifi-cantly higher in smokers than in nonsmokers, which wasinterpreted as due to increased oxidative stress [30]. Elevatedlevels of oxysterols have been detected in tissues in bothalcohol-fed rats and in alcoholic patients [31–33]. Increasedoxysterol levels have also been reported in other disease stateswhere increased oxidative stress is implicated, such as diabetesmellitus and chronic renal failure [5,7].

The two oxysterols usually determined as indices ofoxidative stress are 7β-hydroxycholesterol and 7-ketocholes-terol. These oxysterols are formed by cholesterol autoxidationor secondary to lipid peroxidation [1]. Recently, it was shown

Fig. 5. Appearance of deuterium label in circulating 7-ketocholesterol afterinjection of [2H6]7β-hydroxycholesterol into two healthy volunteers. Withinminutes after infusion the corresponding deuterium label appeared in 8–11% oftotal circulating 7-ketocholesterol.

that 11β-hydroxysteroid dehydrogenase type 1, the enzymeconverting cortisone to cortisol, can metabolize the oxysterols7-ketocholesterol and 7β-hydroxycholesterol. In vitro studiesand studies in experimental animals indicated that 11β-hydroxysteroid dehydrogenase type 1 is an important enzymefor the interconversion of 7-ketocholesterol and 7β-hydro-xycholesterol [18,19]. In the present study we show that thisinterconversion takes place also in vivo in humans. Infusionexperiments with deuterium-labeled oxysterols were carried outto determine the elimination kinetics and as shown in Table 4,7β-hydroxycholesterol and 7-ketocholesterol were rapidlycleared from the circulation with similar half-lives, 1.9 and1.5 h, respectively. These half-lives are comparable to the half-lives of 7α-hydroxycholesterol and 27-hydroxycholesterol butconsiderably shorter than the half-lives of 24-hydroxycholes-terol and 4β-hydroxycholesterol (Table 4). The apparent poolsizes of unlabeled oxysterols were calculated by extrapolatingthe elimination curves to time zero, assuming that the half-lifefor elimination was the same before and after infusion of thedeuterium-labeled steroid. Earlier studies on the elimination ofdeuterium-labeled 4β-hydroxycholesterol in healthy volunteersand in a patient treated with carbamazepine causing a 20-foldincrease in the plasma level of 4β-hydroxycholesterol (and analmost 30-fold increase in apparent pool size) showed almostequal elimination kinetics [34].

We assume that interconversion of 7β-hydroxycholesteroland 7-ketocholesterol is due to the enzyme 11β-hydroxysteroiddehydrogenase type 1, although we have no means of testingthis directly. There is no suitable inhibitor of this enzyme thatcould be administered to humans. Preliminary experiments inour laboratory have failed to demonstrate an effect of liquoriceon the ratio of 7β-hydroxycholesterol to 7-ketocholesterol.Since liquorice has been reported to act mainly on 11β-hydroxysteroid dehydrogenase type 2, no conclusions could bedrawn from the experiments [35]. The interconversion of 7α-and 7β-hydroxydehydroepiandrosterone by the human 11β-hydroxysteroid dehydrogenase type 1 was recently described[36]. Studies with recombinant enzyme showed that both 7α-and 7β-hydroxydehydroepiandrosterone were converted intothe corresponding 7-keto metabolite and that the reversereaction also yielded both isomers. This is in contrast to thesituation with 7-ketocholesterol which is converted stereospe-cifically into 7β-hydroxycholesterol, as shown in this study andby others [18,19].

The conversion of 7-ketocholesterol into 7β-hydroxycho-lesterol apparently accounts for only a small fraction of the

700 H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

metabolism of this oxysterol. Brown and co-workers havepreviously shown that human macrophages metabolize 7-ketocholesterol to 27-hydroxy-7-ketocholesterol by the actionof the enzyme cytochrome P450 27A1 (sterol 27-hydroxylase,CYP27A1) [37]. 7-Ketocholesterol was shown to be metabo-lized to a much greater extent than cholesterol by themacrophages. It has also been shown that a human liver cellline, HepG2, metabolizes 7-ketocholesterol into 27-hydroxy-7-ketocholesterol and experiments with inhibitors suggested thatthe conversion was catalyzed by CYP27A1 [14]. It seems thatthis metabolic pathway is of considerable importance since apatient lacking the sterol 27-hydroxylase had a significantlyincreased concentration of 7-ketocholesterol in plasma [13].

Determination of the intraindividual biological variation inplasma oxysterol concentrations in 12 healthy subjects revealedrelative standard deviations between 20 and 30% for alloxysterols except 24- and 27-hydroxycholesterol. The variationof the latter oxysterols varied only between 8 and 10%, in goodagreement with a recently published study where the oxysterolswere measured by HPLC-MS [38]. The relatively large variationin plasma concentration of 7β-hydroxycholesterol and 7-ketocholesterol with time (Table 2) means that these oxysterolscannot be used for single determinations in individual patients,unless large increases in plasma concentrations are present. Asshown in Table 2, the intraindividual CVs for 7β-hydroxycho-lesterol and 7-ketocholesterol are very similar to the correspond-ing interindividual CVs. The oxysterols may, however, bevaluable markers for comparing groups of patients [39].

In conclusion, the present study clearly shows interconver-sion of infused deuterium-labeled 7β-hydroxycholesterol and7-ketocholesterol in humans. Both oxysterols had similar half-lives in the circulation.

Acknowledgments

This study was supported by grants from AssociazioneItaliana Ricerca sul Cancro, Swedish Heart-Lung Foundation,and Karolinska Institutet Research Foundations.

References

[1] Smith, L. L. Cholesterol autoxidation. New York: Plenum; 1981.[2] Dzeletovic, S.; Babiker, A.; Lund, E.; Diczfalusy, U. Time course of

oxysterol formation during in vitro oxidation of low density lipoprotein.Chem. Phys. Lipids 78:119–128; 1995.

[3] Patel, R. P.; Diczfalusy, U.; Dzeletovic, S.; Wilson, M. T.; Darley-Usmar,V. M. Formation of oxysterols during oxidation of low density lipoproteinby peroxynitrite, myoglobin and copper. J. Lipid Res. 37:2361–2371;1996.

[4] Brown, A. J.; Jessup, W. Oxysterols and atherosclerosis. Atherosclerosis142:1–28; 1998.

[5] Abo, K.; Mio, T.; Sumino, K. Comparative analysis of plasma anderythrocyte 7-ketocholesterol as a marker for oxidative stress in patientswith diabetes mellitus. Clin. Biochem. 33:541–547; 2000.

[6] Murakami, H.; Tamasawa, N.; Matsui, J.; Yasujima, M.; Suda, T. Plasmaoxysterols and tocopherol in patients with diabetes mellitus andhyperlipidemia. Lipids 35:333–338; 2000.

[7] Boaz, M.; Iuliano, L.; Himmelfarb, J.; Matas, Z.; Micheletta, F.;McMonagle, E.; Friedman, V.; Natoli, S.; Gvirtz, G.; Biro, A.; Smetana,S.; Sabo, G.; Gafter, U.; Weinstein, T. Baseline oxysterols and other

markers of oxidative stress, inflammation and malnutrition in the vitamin Eand intima media thickness progression in end-stage renal disease (VIPER)cohort. Nephron Clin. Pract. 100:c111–c119; 2005.

[8] Yoshida, Y.; Niki, E. Detection of lipid peroxidation in vivo: totalhydroxyoctadecadienoic acid and 7-hydroxycholesterol as oxidative stressmarker. Free Radic. Res. 38:787–794; 2004.

[9] Björkhem, I.; Einarsson, K.; Johansson, G. Formation and metabolism of3β-hydroxycholest-5-en-7-one and cholest-5-ene-3β,7β-diol. Acta Chem.Scand. 22:1595–1605; 1968.

[10] Breuer, O.; Sudjana-Sugiaman, E.; Eggertsen, G.; Chiang, J. Y. L.;Björkhem, I. Cholesterol 7α-hydroxylase is up-regulated by the compe-titive inhibitor 7-oxocholesterol in rat liver. Eur. J. Biochem.215:705–710; 1993.

[11] Erickson, S. K.; Cooper, A. D.; Matsui, S. M.; Gould, G. 7-Ketocholesterol. Its effects on hepatic cholesterogenesis and its hepaticmetabolism in vivo and in vitro. J. Biol. Chem. 252:5186–5193; 1977.

[12] Lyons, M. A.; Samman, S.; Gatto, L.; Brown, A. J. Rapid hepaticmetabolism of 7-ketocholesterol in vivo: implications for dietaryoxysterols. J. Lipid Res. 40:1846–1857; 1999.

[13] Burnett, J. R.; Moses, E. A.; Croft, K. D.; Brown, A. J.; Grainger, K.;Vasikaran, S. D.; Leitersdorf, E.; Watts, G. F. Clinical and biochemicalfeatures, molecular diagnosis and long-term management of a case ofcerebrotendinous xanthomatosis. Clin. Chim. Acta 306:63–69; 2001.

[14] Lyons, M. A.; Brown, A. J. Metabolism of an oxysterol, 7-ketocholesterol,by sterol 27-hydroxylase in HepG2 cells. Lipids 36:701–711; 2001.

[15] Song, W.; Pierce, W. M., Jr.; Saeki, Y.; Redinger, R. N.; Prough, R. A. 7α-Hydroxycholesterol oxidoreductase: a novel enzyme found in hamster andhuman liver. FASEB J. 8:A1469; 1994.

[16] Song, W.; Pierce, W. M., Jr.; Saeki, Y.; Refinger, R. N.; Prough, R. A.Endogenous 7-oxocholesterol is an enzymatic product: characterization of7α-hydroxycholesterol dehydrogenase activity of hamster liver micro-somes. Arch. Biochem. Biophys. 328:272–282; 1996.

[17] Song, W.; Chen, J.; Dean, W. L.; Redinger, R. N.; Prough, R. A.Purification and characterization of hamster liver microsomal 7α-hydroxycholesterol dehydrogenase. J. Biol. Chem. 273:16223–16228;1998.

[18] Hult, M.; Elleby, B.; Shafqat, N.; Svensson, S.; Rane, A.; Jörnvall, H.;Abrahmsen, L.; Oppermann, U. Human and rodent type 1 11β-hydroxysteroid dehydrogenases are 7β-hydroxycholesterol dehydro-genases involved in oxysterol metabolism. Cell. Mol. Life Sci. 61:992–999; 2004.

[19] Schweizer, R. A. S.; Zürcher, M.; Balazs, Z.; Dick, B.; Odermatt, A. Rapidhepatic metabolism of 7-ketocholesterol by 11β-hydroxysteroid dehydro-genase type 1. Species-specific differences between the rat, human, andhamster enzyme. J. Biol. Chem. 279:18415–18424; 2004.

[20] Dzeletovic, S.; Breuer, O.; Lund, E.; Diczfalusy, U. Determination ofcholesterol oxidation products in human plasma by isotope dilution - massspectrometry. Anal. Biochem. 225:73–80; 1995.

[21] Russell, D. W. Oxysterol biosynthetic enzymes. Biochim. Biophys. Acta1529:126–135; 2000.

[22] Steinberg, D.; Parthasarathy, S.; Carew, T. E.; Khoo, J. C.; Witztum, J. L.Beyond cholesterol. Modifications of low-density lipoprotein that increaseits atherogenicity. N. Engl. J. Med. 320:915–924; 1989.

[23] Steinberg, D. Low density lipoprotein oxidation and its pathobiologicalsignificance. J. Biol. Chem. 272:20963–20966; 1997.

[24] Witztum, J. L. The oxidation hypothesis of atherosclerosis. Lancet344:793–795; 1994.

[25] Björkhem, I.; Andersson, O.; Diczfalusy, U.; Sevastik, B.; Xiu, R.; Duan,C.; Lund, E. Atherosclerosis and sterol 27-hydroxylase: evidence for a roleof this enzyme in elimination of cholesterol from human macrophages.Proc. Natl. Acad. Sci. USA 91:8592–8596; 1994.

[26] Salonen, J. T.; Nyyssönen, K.; Salonen, R.; Porkkala-Sarataho, E.;Tuomainen, T.-P.; Diczfalusy, U.; Björkhem, I. Lipoprotein oxidation andprogression of carotid atherosclerosis. Circulation 95:840–845; 1997.

[27] Ziedén, B.; Kaminskas, A.; Kristenson, M.; Kucinskiene, Z.; Vessby, B.;Olsson, A. G.; Diczfalusy, U. Increased plasma 7β-hydroxycholesterolconcentrations in a population with a high risk for cardiovascular disease.Arterioscler. Thromb. Vasc. Biol. 19:967–971; 1999.

701H. Larsson et al. / Free Radical Biology & Medicine 43 (2007) 695–701

[28] Porkkala-Sarataho, E.; Salonen, J. T.; Nyyssönen, K.; Kaikkonen, J.;Salonen, R.; Ristonmaa, U.; Diczfalusy, U.; Brigelius-Flohe, R.; Loft, S.;Poulsen, H. E. Long-term effects of vitamin E, vitamin C, and combinedsupplementation on urinary 7-hydro-8-oxo-2′-deoxyguanosine, serumcholesterol oxidation products, and oxidation resistance of lipids innondepleted men. Arterioscler. Thromb. Vasc. Biol. 20:2087–2093; 2000.

[29] Lykkesfeldt, J.; Viscovich, M.; Poulsen, H. E. Plasma malondialdehyde isinduced by smoking: a study with balanced antioxidant profiles. Br. J.Nutr. 92:203–206; 2004.

[30] Mol, M. J. T. M.; de Rijke, Y. B.; Demacker, P. N. M.; Stalehoef, A. F. H.Plasma levels of lipid and cholesterol oxidation products and cytokines indiabetes mellitus and cigarette smoking: effects of vitamin E treatment.Atherosclerosis 129:169–176; 1997.

[31] Adachi, J.; Kudo, R.; Ueno, Y.; Hunter, R.; Rajendram, R.; Want, E.;Preedy, V. R. Heart 7-hydroperoxycholesterol and oxysterols are elevatedin chronically ethanol-fed rats. J. Nutr. 131:2916–2920; 2001.

[32] Adachi, J.; Hojo, K.; Ueno, Y.; Naito, T.; Ninomiya, I.; Imamichi, H.;Tatsuno, Y. Identification of cholesta-3,5-dien-7-one by gas chromato-graphy-mass spectrometry in the erythrocyte membrane of alcoholicpatients. Alcohol. Clin. Exp. Res. 20:51A–55A; 1996.

[33] Adachi, J.; Kudo, R.; Asano, M.; Ueno, Y.; Hunter, R.; Rajendram, R.;Martin, C.; Preedy, V. R. Skeletal muscle and liver oxysterols duringfasting and alcohol exposure. Metabolism 55:119–127; 2006.

[34] Bodin, K.; Andersson, U.; Rystedt, E.; Ellis, E.; Norlin, M.; Pikuleva, I.;Eggertsen, G.; Björkhem, I.; Diczfalusy, U. Metabolism of 4β-hydroxy-cholesterol in humans. J. Biol. Chem. 277:31534–31540; 2002.

[35] Stewart, P. M.; Wallace, A. M.; Valentino, R.; Burt, D.; Shackleton, C. L.;Edwards, C. R. W. Mineralocorticoid activity of liquorice: 11-beta-hydroxysteroid dehydrogenase deficiency comes of age. Lancet 2:821–824; 1987.

[36] Muller, C.; Pompon, D.; Urban, P.; Morfin, R. Inter-conversion of 7α-and 7β-hydroxy-dehydroepiandrosterone by the human 11β-hydroxys-teroid dehydrogenase type 1. J. Steroid Biochem. Mol. Biol. 99:215–222;2006.

[37] Brown, A. J.; Watts, G. F.; Burnett, J. R.; Dean, R. T.; Jessup, W. Sterol 27-hydroxylase acts on 7-ketocholesterol in human atherosclerotic lesions andmacrophages in culture. J. Biol. Chem. 275:27627–27633; 2000.

[38] Burkard, I.; Von Eckardstein, A.; Waeber, G.; Vollenweider, P.; Rentsch, K. M.Lipoprotein distribution and biological variation of 24S- and 27-hydroxycho-lesterol in healthy volunteers. Atherosclerosis, doi:10.1016/j.athersclerosis.2006.09.026; 2006.

[39] Arca, M.; Natoli, S.; Micheletta, F.; Riggi, S.; Di Angelantonio, E.;Montali, A.; Antonini, T. M.; Antonini, R.; Diczfalusy, U.; Iuliano, L.Increased plasma levels of oxysterols, in vivo markers of oxidative stress,in patients with familial combined hyperlipidemia (FCHL): reductionduring atorvastatin and fenofibrate therapy. Free Radic. Biol. Med.42:698–705; 2007.

[40] Meaney, S.; Hassan, M.; Sakinis, A.; Lütjohann, D.; von Bergmann, K.;Wennmalm, Å.; Diczfalusy, U.; Björkhem, I. Evidence that the majoroxysterols in human circulation originate from distinct pools ofcholesterol: a stable isotope study. J. Lipid Res. 42:70–78; 2001.

[41] Babiker, A.; Andersson, O.; Lindblom, D.; van der Linden, J.; Wiklund,B.; Lütjohann, D.; Diczfalusy, U.; Björkhem, I. Elimination of cholesterolas cholestenoic acid in human lung by sterol 27-hydroxylase: evidence thatmost of this steroid in the circulation is of pulmonary origin. J. Lipid Res.40:1417–1425; 1999.

[42] Björkhem, I.; Lütjohann, D.; Diczfalusy, U.; Ståhle, L.; Ahlborg, G.;Wahren, J. Cholesterol homeostasis in human brain: turnover of 24S-hydroxycholesterol and evidence for a cerebral origin of most of thisoxysterol in the circulation. J. Lipid Res. 39:1594–1600; 1998.